I decided to repost one of my original blog posts on how to easily build and test your connection strings that was quite popular and somehow lost during the transition to the new site.  This method is fairly simple and I use it all the time when I need to build or test out a new connection string.  The first thing you need to do is create a new empty file and call it Test.udl.  Normally I use Notepad to do this, but you can use anything you want.

Once you have done this your file will display the computer icon associated with .udl files and if you double click on the file it will open up the Data Link Properties window.

Next, you need to fill out your connection information including provider, server, database, and login information.  In order to properly get your connection string you will need to check the allow saving password box.  At this point you can click the test connection button to verify that the connection is successful.  If it is not, you can continue to make adjustments to the Data Link Properties window until it is.

Finally, click OK to save the file, and open up the .udl file using Notepad.  If you did check the allow saving password box you will be cautioned about the security risk of saving the password.  Once you’ve opened the file you will see something similar to the follow:

[oledb]

; Everything after this line is an OLE DB initstring

Provider=SQLOLEDB.1;Password=yourpassword;Persist Security Info=True;User ID=sa;Initial Catalog=TimeTracker;Data Source=WDMPDC01

You have now created your connection string and can copy and paste this into your application.  This is also an excellent way of testing connectivity issues from individual computers.

Did you find this posting helpful?  Please let me know by posting a comment.

Scott Hanselman hit the nail on the head with his recent post.  It gave me quite the laugh, but unfortunately it is so true (Just like when I watch Office Space).  I was told the other day by a coworker that I over engineered a solution because I used an httphandler instead of just making a page.  Anyone with a small understanding of ASP.NET should realize that using an httphandler when no content needs to be display is going to be more efficient.  This “page” was going to be hit about every time someone accessed the portal and from what I have come across using an httphandler will increase performance by 5%-10%.  Considering the development time was only increased by the one minute or so that it took to add the proper section to the web.config I would have to say that in this scenario the httphandler was the best way to go and I think that the solution could only be considered “over engineered” if you don’t understand it.  Anyways, I’m beginning to think there is a case of ADD going on in my new environment.

Like I said in an earlier post I’m currently working on a side project with a couple former coworkers that I hope with become a success.  This project has turned my dining room table into an ad hoc office several evening each week and caused the need to setup a development infrastructure to support the project with the backbone of that being the source control repository.  I decided to use Subversion for the source control repository for the initial reason that it was free, but also because it seems to be fairly highly regarded in the community.  I’ve used both Visual SourceSafe 6.0 and 2005 in the past and whereas I’ve never had any direct complaints with VSS it did always seem to be a bit of a task getting things setup initially when starting a project.

Subversion is working fairly well for us.   Because we cannot always get together to work on the project I have http access setup and my partners can very easily sync their source code over the Internet.  Subversion integrates nicely with Apache web server to do this.  Let me point out though that I have all this running on Windows, not Linux.  There are Windows versions of both Subversion and Apache.  For the client side of things we are using both TortoiseSVN and AnkhSvn.  TortoiseSVN integrates directly into Explorer and is handy for setting up new projects and handling the tasks at the file system level.  AnkhSvn is an add-on to Visual Studio that integrates Subversion tasks into the right-click menu of Solution Explorer.

Anyways my goal in this post is to help direct anyone who is Googling trying to setup Subversion get going in the right direct.  Ralph Willgos put together a great article on Code Project walking through the steps of getting everything setup so I will not bother rewriting them.  It is a little old and I skipped the steps of setting up IIS and Visual Studio 2003, but the rest of the steps apply.   I will caution you on a couple things.  First make sure you are using the 2.0.x version of Apache, not 2.2.x and not 1.x.  From my understanding and experience Subversion will only work with the 2.0.x version.  The other thing I will caution you on is to not use the download links that Ralph provides.  They have become out of date as newer versions have been released.  I ran into very cryptic error messages when I had different versions of Subversion and TortoiseSVN trying to talk to each other and this caused several headaches.  To get the files simply go to the links that I have provided and grab the latest versions.

That’s about it.  Feel free to drop me a line if you get stuck with Subversion and I will try to help you out.  Also, I’d like to read what everyone else has to say about source control in general and what they think about specific products so feel free to post a comment with your experiences.  I’m also running a continuous integration server for our project, but I will go into that in a later post.